Proposed here is a population-based case-control study of the role of alcohol consumption in the etiology of cancer of the lung. Lifetime alcohol consumption including frequency, quantity per occasion, beverage type and setting of drinking will all be evaluated with control for potential confounders: smoking (lifetime history of amount, frequency and type), diet (carotenoids, fat, cholesterol and total kilocalorie intake), passive smoke exposure (lifetime exposure at home, work an socializing) and occupational history. In addition, interaction of alcohol with smoking and with diet in risk of lung cancer of the lung in Erie and Niagara counties will be interviewed (560 white males, 500 white females, 165 black males and 65 black females). Controls will be randomly selected, those under age 65 from driver's license enrollees; and those 65 and over from enrollment lists for the Health Care Finance Administration, all frequency matched to cases on age, sex, race and county of residence. For those under age 69, controls will also be used in the comparisons for the case-control study of myocardial infarction. For all cases there will be least two controls. Interview data and blood samples form controls will be used to evaluate the relationships between the comprehensive measures of alcohol use and smoking, passive smoking, diet, occupation, and blood measures (carotenoids, selenium, RBC glutathione peroxidase, lipid peroxidase (MDA), vitamins A, C, and E, total cholesterol, LDL-cholesterol and HDL-cholesterol), in a sample of the general population and by race-and sex-specific categories. This study provides an important opportunity for an efficient examination of alcohol and clarification of the etiology of this significant public health problem. Additionally, data on distribution of alcohol and other risk factors in the general population will add to our understanding of alcohol as a risk factor in chronic disease. These questions can be addressed in a valid and economical way by combining several specific aims in this component and the research component on myocardial infarction and the combined use of the controls in the two research components.